Day-to-Day Variations in Fasting Plasma Glucose Do Not Influence Gastric Emptying in Subjects With Type 1 Diabetes

OBJECTIVE Acute experimental variations in glycemia decelerate (hyperglycemia) or accelerate (hypoglycemia) gastric emptying. Whether spontaneous variations in fasting plasma glucose (FPG) have a similar influence on gastric emptying is yet unclear. RESEARCH DESIGN AND METHODS Gastric emptying of a mixed meal was prospectively studied three times in 20 patients with type 1 diabetes and 10 healthy subjects with normal glucose tolerance using a 13C-CO2 octanoate breath test with Wagner-Nelson analysis. The velocity of gastric emptying was related to FPG measured before the test (grouped as low, intermediate, or high). In addition, gastric emptying data from 255 patients with type 1 diabetes studied for clinical indications were compared by tertiles of baseline FPG. RESULTS Despite marked variations in FPG (by 4.8 [95% CI 3.4; 6.2] mmol/L), gastric emptying did not differ among the three prospective examinations in patients with type 1 diabetes (Δ T1/2 between highest and lowest FPG: 1 [95% CI −35; 37] min; P = 0.90). The coefficient of variation for T1/2 determined three times was 21.0%. Similar results at much lower variations in FPG were found in healthy subjects. In the cross-sectional analysis, gastric emptying did not differ between the tertiles of FPG (Δ T1/2 between highest and lowest FPG: 7 [95% CI −10; 23] min; P = 0.66), when FPG varied by 7.2 (6.7; 7.8) mmol/L. However, higher HbA1c was significantly related to slower gastric emptying. CONCLUSIONS Day-to-day variations in FPG not induced by therapeutic measures do not influence gastric emptying significantly. These findings are in contrast with those obtained after rapidly clamping plasma glucose in the hyper- or hypoglycemic concentrations range and challenge the clinical importance of short-term glucose fluctuations for gastric emptying in patients with type 1 diabetes. Rather, chronic hyperglycemia is associated with slowed gastric emptying.